Transistors and semiconductors

In recent years the transistor — an entirely new type of electron device — has come into its own and bids to replace the bulky electron tubes in many applications. Transistors are far smaller than tubes, have no filament and hence need no heating power. They are mechanically rugged, have practically unlimited life, and can do some jobs better than electron tubes, while catching up fast in other respects.

In contrast to electron tubes, which utilize the flow of free electrons through a vacuum or gas, the transistor relies for its operation on the movement of charge carriers through a solid substance, a semiconductor. Transistors are only one of the family of semiconductors; many other semiconductor applications are becoming increasingly popular and new ones are constantly being discovered.

It is known that materials are classed as semiconductors if their electrical conductivity is intermediate between metallic conductors, which have a large number of free electrons available as charge carriers, and non-metallic insulators, which have practically no free electrons available to conduct current. The two semiconductors most frequently used in electronics and transistor manufacture are germanium and silicon. Both elements have the same crystal structure and similar characteristics, so that the discussion that follows for germanium will also apply to silicon.

It is known that outermost electron shell of an atom contains the loosely held valence electrons, which are easily dislodged to become electric current carriers. Germanium has four valence electrons in its outer shell, and for our purposes, the atom may be pictured as containing only these electrons and four protons in the nucleus to keep it electrically neutral.

When germanium is in crystalline form its atoms assume the typical diamond structure. In this structure adjacent germanium atoms share their valence electrons in a strong bond, so that effectively four orbital electron pairs are associated with each nucleus. These electron pairs are termed covalent bonds and they are bound so strongly to each other and to the nucleus that no free electrons are available to conduct a current through the germanium.

A pure germanium crystal, therefore, is practically a non-conductor of electricity. It is not completely non-conducting, since ordinary heat energy occasionally disrupts some of the covalent bonds, thus liberating free electrons as charge carriers.

If a small amount of an impurity is introduced into the germanium crystal, its current-conducting characteristics change radically. Thus, when atoms that have five electrons in their outer shell are introduced into the germanium, a procedure known as doping, the fifth electron of the impurity atom does not find a place in thesymmetrical covalent-bond structure and, hence, is free to roam around through the crystal. These free electrons are then available as electric current carriers.

By placing an electric field across the "doped" germanium crystal, the excess of free electrons donated by the impurity atoms will travel toward the positive terminal of the voltage source. Relatively few impurity or "donor" atoms within thegermanium structure permit fairly substantial electron currents through the crystal when an electric field is applied. Germanium that has been doped by pen-tavalent donor atoms is known to be n-type germanium, because current conduction is carried on with negative charge carriers, or electrons.

Consider now the situation when an impurity that has only three electrons in its outer shell is introduced into the pure germanium crystal. The trivalent indium atoms take their place in the germanium structure, but one of the covalent bonds around each indium atom has an electron missing, or a hole in its place. Although the hole indicates the absence of an electron it behaves like a real, positively charged particle when an electric field is applied across the crystal.

Under the influence of the electric field, electrons within the crystal will tend to move toward the positive terminal of the voltage source and jump into the available holes of the indium atoms near the positive terminal. Since there are no free electrons available, the deficient indium atoms near the positive terminal "steal" electrons from their neighbors by disrupting their covalent bonds. This creates new holes in adjacent atoms.

As electrons move toward the positive terminal, the j holes will move toward the negative terminal, thus acting like mobile, positive particles. As the holes reach the negative terminal, electrons enter the crystal near the terminal and combine with the holes, thus cancelling them.

At the same time, the loosely held electrons that filled the holes near the positive terminal, are attracted away from their atoms into the positive terminal. This, of course, creates new holes near the positive terminal, which again drift toward the negative terminal. Current conduction may thus be considered to occur by means of holes inside the crystal, and by means of electrons through the external connecting wires and battery.

An impurity that has three electrons in its outer shell is known as an acceptor atom, because it takes electrons away from surrounding germanium atoms. Germanium that has been doped with trivalent acceptor atoms is called p-type germanium, to specify that current conduction is carried on by holes, which are the equivalent of positive charges.